I don't think the problem is string theory per se — I think the problem is that we're running up on the bounds of what we can test experimentally, and further progress is going to require some really cleverly designed experiments that can probe Planck scale events. Although the greatest physicists throughout history have for the most part been theoreticians, it may be the case that in the future, the greatest physicists will be experimentalists because of the extraordinary amount of creativity needed to design novel experiments that can probe the most obscure corners of reality.

String theory isn't pseudoscience, because it does make testable, falsifiable predictions. In fact, it makes the same predictions that quantum field theory does for the phenomena that we are currently capable of testing (and arguably using a more elegant framework than QFT provides). The problem is that the predictions that string theory makes beyond QFT are currently way outside the realm of experimental assessment. This doesn't make the predictions false — if that's the way the universe really is, we just may be incapable of knowing that for a long time. There are a few hints that at least some of the predictions of string theory may be incorrect — we have yet to find any evidence of supersymmetry, and many physicists thought the LHC would turn up at least some evidence of this if it existed.

Regardless, I think people get too hung up on buzz-phrases like "is time real?" or "are we in a multiverse?". A lot of these are what I consider irrelevant to science, as the goal of science is to make accurate predictions about the future. If there is another universe out there, it is by definition unreachable from our current universe, so it makes no difference to science whatsoever. Likewise, what is or isn't "real" has no bearing on how accurately one can compute the evolution of some system's state through time.